universal-timeseries-transformer 0.2.5

A package for transforming and manipulating time series data with universal interfaces

Universal Timeseries Transformer

A Python package that provides a universal interface for transforming and manipulating time series data. This package offers flexible and efficient tools for handling various types of time series data transformations.

Version Updates

v0.2.5 (2025-06-15)

• Added timeseries_splitter module for splitting timeseries data into two-columned format
• Fixed incomplete function in timeseries_splitter module

v0.2.4 (2025-06-08)

• Modified return calculation functions to display returns in percentage format (multiplied by 100)
• Updated all return-related functions in timeseries_application.py

v0.2.3 (2025-06-04)

• Added new properties to PricesMatrix class: ytd_date_pairs, date_inception, date_end
• Updated string_date_controller dependency to version 0.2.3 or higher

v0.2.2 (2025-06-04)

• Enhanced exception handling in PricesMatrix class
• Added set_date_ref method for better date reference management

v0.2.1 (2025-06-03)

• Added monthly_date_pairs property to PricesMatrix class for convenient monthly date analysis
• Updated string_date_controller dependency to version 0.2.1 or higher

v0.2.0 (2025-06-03)

• Major version update as the module reaches maturity
• Added date_ref property to PricesMatrix class for improved date reference handling
• All features from previous versions are now stable and production-ready

v0.1.10 (2025-06-03)

• Fixed bug in PricesMatrix class to use correct string_date_controller function
• Updated to use get_all_data_historical_dates function from string_date_controller 0.2.0

v0.1.9 (2025-06-02)

• Fixed bug in PricesMatrix class related to historical dates calculation
• Updated to use correct string_date_controller functions

v0.1.8 (2025-06-02)

• Added PricesMatrix class extending TimeseriesMatrix for price data handling
• Enhanced matrix representation capabilities with historical dates support

v0.1.7 (2025-06-02)

• Improved TimeseriesMatrix class with optimized property handling
• Updated string_date_controller dependency to version 0.2.0 or higher
• Removed unused date_calculus module

v0.1.6 (2025-06-01)

• Added timeseries_slicer module with date-based and index-based slicing functions
• Added timeseries_extender module with enhanced date extension functionality
• Improved .gitignore to exclude Jupyter notebook files

v0.1.5 (2025-05-30)

• Added TimeseriesMatrix class for matrix representation of time series data
• Enhanced data access with row, column, and component selection methods
• Added format conversion methods (datetime, unixtime, string)

v0.1.4 (2025-05-28)

• Added verbose option to control log output
• Enhanced timeseries extension functionality
• Improved code readability and documentation

v0.1.3 (2025-05-19)

• Added new timeseries_application module with financial calculations
• Added functions for returns and cumulative returns calculation

v0.1.2 (2025-05-19)

• Improved stability and performance optimization
• Enhanced type checking functionality
• Documentation improvements

Features

• Index Transformer
  • Flexible time index manipulation
  • Date range operations
  • Frequency conversion
• DataFrame Transformer
  • Universal interface for time series operations
  • Data alignment and merging
  • Efficient data transformation
• Timeseries Basis
  • Core functionality for time series manipulation
  • Common time series operations

Installation

You can install the package using pip:

pip install universal-timeseries-transformer

Requirements

• Python >= 3.8
• Dependencies:
  • pandas
  • numpy

Usage Examples

1. Basic Time Series Transformation

from universal_timeseries_transformer import IndexTransformer, DataFrameTransformer
import pandas as pd

# Create sample time series data
df = pd.DataFrame({'value': [1, 2, 3, 4]},
                  index=pd.date_range('2025-01-01', periods=4))

# Transform time series index
index_transformer = IndexTransformer(df)
weekly_data = index_transformer.to_weekly()

# Apply data transformations
df_transformer = DataFrameTransformer(weekly_data)
result = df_transformer.rolling_mean(window=2)

2. Advanced Time Series Operations

from universal_timeseries_transformer import TimeseriesBasis

# Initialize time series basis
ts_basis = TimeseriesBasis(df)

# Perform complex transformations
transformed_data = ts_basis.transform()

)

Find funds with borrowings

funds_with_borrowings = search_funds_having_borrowings(date_ref='2025-02-21')

Get borrowing details

fund_code = '100075' borrowing_details = get_borriwings_by_fund(fund_code=fund_code, date_ref='2025-02-21')

### 3. Check Repo Agreements

```python
from financial_dataset_preprocessor import (
    search_funds_having_repos,
    get_repos_by_fund
)

# Find funds with repos
funds_with_repos = search_funds_having_repos(date_ref='2025-02-21')

# Get repo details for a specific fund
fund_code = '100075'
repo_details = get_repos_by_fund(fund_code=fund_code, date_ref='2025-02-21')

Development

To set up the development environment:

1. Clone the repository
2. Create a virtual environment
3. Install dependencies:

pip install -r requirements.txt

License

This project is licensed under a proprietary license. All rights reserved.

Terms of Use

• Source code viewing and forking is allowed
• Commercial use is prohibited without explicit permission
• Redistribution or modification of the code is prohibited
• Academic and research use is allowed with proper attribution

Author

June Young Park
AI Management Development Team Lead & Quant Strategist at LIFE Asset Management

LIFE Asset Management is a hedge fund management firm that integrates value investing and engagement strategies with quantitative approaches and financial technology, headquartered in Seoul, South Korea.

Contact

• Email: [e-mail]
• Location: TWO IFC, Yeouido, Seoul